GenUP: Generative User Profilers as In-Context Learners for Next POI Recommender Systems

2024-10-28

arXiv.org

AI Summary

To address the three key challenges in POI recommendation—lack of model transparency (black-box nature), poor cold-start robustness, and limited scalability—this paper proposes a lightweight large language model (LLM) in-context learning framework grounded in natural-language (NL) user profiling. Departing from conventional vector-based embeddings and trajectory-reuse paradigms, we introduce the first NL profiling method that integrates psychological personality assessment with behavioral theory, transforming LBSN check-in data into interpretable, incrementally updatable textual user representations. These NL profiles serve as structured prompts for LLMs to perform next-POI prediction. The approach ensures interpretability, cold-start adaptability, and computational efficiency. Extensive experiments on multiple real-world datasets demonstrate significant improvements over state-of-the-art methods: notably higher recommendation accuracy, 57% reduction in inference overhead, real-time profile updating capability, and plug-and-play deployment feasibility.

Technology Category

• Natural Language Processing: (Large) Language Models
• Machine Learning: Learning Preferences or Rankings
• Data Mining & Knowledge Management: Conversational Systems for Recommendation & Retrieval

Application Category

• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

Traditional Point-of-Interest (POI) recommendation systems often lack transparency, interpretability, and scrutability due to their reliance on dense vector-based user embeddings. Furthermore, the cold-start problem -- where systems have insufficient data for new users -- limits their ability to generate accurate recommendations. Existing methods often address this by leveraging similar trajectories from other users, but this approach can be computationally expensive and increases the context length for LLM-based methods, making them difficult to scale. To address these limitations, we propose a method that generates natural language (NL) user profiles from large-scale, location-based social network (LBSN) check-ins, utilizing robust personality assessments and behavioral theories. These NL profiles capture user preferences, routines, and behaviors, improving POI prediction accuracy while offering enhanced transparency. By incorporating NL profiles as system prompts to LLMs, our approach reduces reliance on extensive historical data, while remaining flexible, easily updated, and computationally efficient. Our method is not only competitive with other LLM-based and complex agentic frameworks but is also more scalable for real-world POI recommender systems. Results demonstrate that our approach consistently outperforms baseline methods, offering a more interpretable and resource-efficient solution for POI recommendation systems. Our source code is available at: https://github.com/w11wo/GenUP/.

Problem

• Addresses lack of transparency in POI recommendation systems
• Solves cold-start problem for new users in POI recommendations
• Reduces computational cost and improves scalability in LLM-based methods

Innovation

• Generates natural language user profiles
• Utilizes personality assessments and behavioral theories
• Incorporates NL profiles as LLM system prompts